Icing thickness calculation method and system based on continuous weighing and inclination angle data

Abstract

The invention discloses an icing thickness calculation method and system based on continuous weighing and inclination angle data. The method comprises the following steps: acquiring first task data of a first object, and acquiring second task data of a task area of the first object; and performing mutual verification by using the first task data and the second task data, and performing first preprocessing on the data. Environment data are collected, and a data set is constructed by combining the environment data with the preprocessed data. And constructing a neural network model, and completing task calculation. According to the method, abnormal values are effectively eliminated, and the data standardization level is improved, so that the data accuracy and reliability are remarkably improved. And the accuracy and robustness of a calculation result are improved by utilizing the supplementary effect of meteorological data on icing thickness prediction. Through training and optimization of the model, efficient prediction of the icing thickness of the power transmission line is realized, and an intelligent means is provided for real-time monitoring.

Description

Technical Field

[0001] The present invention relates to the technical field of transmission line icing monitoring, and in particular to an icing thickness calculation method and system based on continuous weighing and inclination data. Background Art

[0002] With the continuous development of the power system, the operational reliability of transmission lines under various extreme climatic conditions has become a key research issue. In cold winters or high-altitude areas, icing of transmission lines often leads to weight gain, tension changes, and damage to mechanical structures. In severe cases, it may cause accidents such as line breakage or tower collapse. Traditional icing monitoring methods mainly rely on manual inspections and limited single-point sensor detection, but the frequency and accuracy of manual inspections are low, making it difficult to detect the severity of line icing in a timely manner. Although single-point sensors can provide local data, they cannot achieve real-time and comprehensive monitoring of ice thickness over a large area, and the data is not representative and comprehensive enough. In addition, meteorological factors such as temperature, humidity, and wind speed have a complex impact on ice formation. Traditional methods have a low utilization rate of meteorological data and cannot fully explore the potential correlation between meteorological data and ice thickness.

[0003] The existing ice monitoring technology based on weighing and inclination sensors has made some progress. By sensing the weight change and inclination of the line, the ice thickness estimation has been preliminarily realized. However, these methods have the following main shortcomings: First, the data processing accuracy is low. Due to the complex sensor installation environment, the collected data may contain noise and outliers. The existing methods lack effective processing methods in data cleaning and standardization, resulting in limited accuracy of the calculation results; second, there is a lack of fusion capabilities of multiple data sources. Existing systems often rely solely on weighing or inclination data, and fail to fully utilize the complementarity of the two and the synergy of meteorological data; third, the model capabilities are insufficient. Most of the current ice calculations use simple regression or empirical models, lack deep modeling capabilities for complex multivariate time series data, and find it difficult to capture nonlinear relationships and time dynamic characteristics between variables. Summary of the invention

[0004] In view of the above-mentioned problems, the present invention is proposed.

[0005] Therefore, the technical problem solved by the present invention is: the existing ice monitoring technology based on weighing and inclination sensors has made certain progress. By sensing the weight change and inclination of the line, the ice thickness estimation has been preliminarily realized. However, these methods mainly have the following shortcomings: First, the data processing accuracy is low. Due to the complex sensor installation environment, the collected data may contain noise and outliers. The existing methods lack effective processing means in data cleaning and standardization, resulting in limited accuracy of the calculation results; second, the lack of fusion capabilities of multiple data sources. Existing systems often rely solely on weighing or inclination data, and fail to fully utilize the complementarity of the two and the synergy of meteorological data; third, the model capabilities are insufficient. Most of the current ice calculations use simple regression or empirical models, lack deep modeling capabilities for complex multivariate time series data, and find it difficult to capture nonlinear relationships and time dynamic characteristics between variables.

[0006] To solve the above technical problems, the present invention provides the following technical solutions: a method for calculating ice thickness based on continuous weighing and inclination data, comprising: collecting first task data of a first object, and collecting second task data of a task area of ​​the first object.

[0007] The first task data and the second task data are used to perform mutual verification and perform a first pre-processing on the data.

[0008] Collect environmental data and build a data set by combining it with pre-processed data.

[0009] Build a neural network model to complete the calculation of the task.

[0010] As a preferred solution of the ice thickness calculation method based on continuous weighing and inclination data described in the present invention, the first task data includes installing a collection device in the first object to collect the first task data.

[0011] As a preferred solution of the ice thickness calculation method based on continuous weighing and inclination data described in the present invention, wherein: the mission area of ​​the first object is the ice-covered area of ​​the first object, and a collection device is installed to collect second mission data.

[0012] As a preferred solution of the ice thickness calculation method based on continuous weighing and inclination data described in the present invention, the mutual verification includes comparing and analyzing the first task data and the second task data, verifying the consistency of the data, and deleting inconsistent data.

[0013] The first pre-processing is to pre-process the retained consistent data to improve data accuracy and standardization.

[0014] As a preferred solution of the ice thickness calculation method based on continuous weighing and inclination data described in the present invention, the constructing of the data set includes integrating the first environmental data and the pre-processed data to construct the data set.

[0015] As a preferred solution of the ice thickness calculation method based on continuous weighing and inclination data of the present invention, wherein:

[0016] As a preferred solution of the ice thickness calculation method based on continuous weighing and inclination data described in the present invention, wherein: the first object includes but is not limited to a transmission line, and the first task data includes ice weight and ice thickness marking data on the first object.

[0017] The second mission data includes the inclination data of the mission area of ​​the first object and the ice thickness mark data.

[0018] The first pre-processing includes but is not limited to data cleaning and standardization.

[0019] Environmental data includes meteorological data.

[0020] An ice thickness calculation system based on continuous weighing and inclination data, characterized in that it includes:

[0021] The constructing of the neural network model to complete the calculation of the first task includes constructing a deep learning network and training the model using the data set.

[0022] Based on the trained neural network model, the newly collected first task data and second task data are input to complete the calculation of the task.

[0023] A computer device comprises a memory and a processor, wherein the memory stores a computer program, and the processor implements the steps of the above method when executing the computer program.

[0024] A computer-readable storage medium stores a computer program, which implements the steps of the method described above when executed by a processor.

[0025] The beneficial effects of the present invention are as follows: by collecting the weighing data and inclination data of the first object, and mutually verifying the two types of data, preprocessing is performed after screening out data with high consistency, effectively eliminating outliers and improving the level of data standardization, thereby significantly improving the accuracy and reliability of the data.

[0026] The meteorological data and the processed weighing and inclination data are integrated into a multivariate data set. The complex correlations between variables are mined through a deep learning model, and the supplementary role of meteorological data in ice thickness prediction is fully utilized to improve the accuracy and robustness of the calculation results.

[0027] A deep learning network is designed to adapt to the multivariate time series data characteristics of ice monitoring. Through model training and optimization, efficient prediction of ice thickness on transmission lines is achieved, providing an intelligent means for real-time monitoring.

[0028] The present invention is applicable to a variety of scenarios, including key locations with complex icing conditions such as power transmission lines, main cables of suspension bridges and wind turbine blades. The system is versatile and extensible, and can adjust data collection and processing parameters according to specific application requirements to meet monitoring needs in different scenarios. BRIEF DESCRIPTION OF THE DRAWINGS

[0029] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for describing the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work. Among them:

[0030] Figure 1 An overall flow chart of an ice thickness calculation method and system based on continuous weighing and inclination data provided in the first embodiment of the present invention. DETAILED DESCRIPTION

[0031] In order to make the above-mentioned purposes, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are described in detail below in conjunction with the drawings of the specification. Obviously, the described embodiments are part of the embodiments of the present invention, but not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary persons in the art without creative work should fall within the scope of protection of the present invention.

[0032] Example 1, reference Figure 1 , as an embodiment of the present invention, provides a method for calculating ice thickness based on continuous weighing and inclination data, comprising:

[0033] S1: Collecting first task data of a first object and collecting second task data of a task area of ​​the first object.

[0034] In the present invention, the first object is the distribution line, and the first task data is to use a continuous weighing device to monitor the ice weight and ice thickness mark on the distribution line in real time. The task area of ​​the first object is the ice-covered area line in the distribution line, and the second task data is to collect the inclination data and ice thickness mark data of the ice-covered area line using an inclination sensor.

[0035] Collecting the data for the first task includes selecting a suitable continuous weighing device to ensure that it has the characteristics of high precision, high reliability, and high stability. Installing the continuous weighing device on the side of the tower to ensure its stable installation and avoid errors caused by improper installation. Configuring the data transmission system to ensure the real-time, reliability, and security of the data. Calibrating the continuous weighing device to ensure its accuracy and stability. Collecting data and transmitting the data to the data processing center.

[0036] Collecting the data for the second mission includes selecting appropriate inclination sensors to ensure high accuracy, stability, and adaptability. Accurately installing inclination sensors on the lines in the ice-covered area to ensure the accuracy and reliability of data collection. Calibrating the inclination sensors to ensure the accuracy and comparability of the data. Collecting inclination data and ice thickness marker data in real time, and ensuring accurate transmission and recording of the data. Performing preliminary processing on the collected data, such as removing outliers and noise.

[0037] Draw a box plot for each numerical variable in the data set. The box plot shows the data distribution through quartiles and interquartile ranges. Outliers are usually located outside the box.

[0038] According to the quartiles and interquartile range (IQR) of the box plot, the threshold of the outlier is determined, which is generally located outside the lower quartile minus 1.5 times the IQR or the upper quartile plus 1.5 times the IQR. Analyze the identified outliers to determine whether they are real data variations or caused by errors. Decide on the processing strategy: If the outlier is a real variation, keep it. If it is caused by an error, choose to delete it or replace it with the mean, median, K-nearest neighbor, etc.

[0039] Re-draw the box plot to verify the processing results and ensure that the data distribution characteristics have not been changed by mistake. Record each step and decision of outlier processing for review and audit.

[0040] It should be noted that the first object includes but is not limited to a power distribution line, and may also be a main cable of a large suspension bridge or a blade structure of a wind turbine generator set.

[0041] In an optional embodiment of the present invention, the first object is the main cable of a large suspension bridge. Under severe cold and humid climate conditions, ice is easily attached to the surface of the main cable in winter or extreme climate, affecting the deadweight and stress distribution of the main cable, which may have a potential impact on the overall safety and stability of the bridge. The first task is to measure and monitor the ice thickness of the main cable. In this scenario, the task is to obtain the thickness distribution of the ice layer on the main cable by detecting the weight change and tilt change of the main cable caused by ice coating, so as to facilitate bridge maintenance personnel to perform de-icing or other maintenance measures when necessary.

[0042] The first task data includes:

[0043] Continuous weighing data: A high-precision continuous weighing sensor is installed at the anchor end of the main cable or a specific cable section to detect the change in the deadweight of the main cable with or without ice coverage, and obtain information related to the additional mass of the ice layer.

[0044] Ice thickness labeling data: In the early test stage, the ice thickness at certain time points can be calibrated and recorded manually or using reference equipment. These labeling data are used as reference in subsequent model training.

[0045] Mission area of ​​the first object:

[0046] The key areas of the main cable that are most prone to ice accumulation (such as the mid-span section or the section close to the densely distributed suspension cables) are where weighing devices and inclination sensors are installed simultaneously.

[0047] Second task data:

[0048] Inclination data: High-precision inclination sensors are placed at specific locations on the main cable to monitor the impact of uneven loads caused by ice on the slight inclination and strain of the main cable at any time. Inclination data and weighing data are mutually verified, which can filter out abnormal values ​​in data preprocessing and improve data reliability.

[0049] The specific collection steps include installing a continuous weighing device at the main cable anchor point or a specific cable section, and calibrating the device to ensure that it can monitor load changes stably over a long period of time. Install an inclination sensor at specific nodes of the main cable (such as the mid-span or key section) to ensure that the tiny displacement and tilt changes of the main cable caused by ice can be recorded in real time. The weighing data and inclination data are transmitted to the data processing center in real time via optical fiber, wireless network or wired network to ensure data security and integrity. Match the initial manually calibrated ice thickness information with the real-time data of the equipment to establish a basic data set for training.

[0050] In an optional embodiment of the present invention, the first object is the blade structure of a wind turbine. In a cold and humid environment, ice is easily formed on the surface of the wind turbine blades, affecting the aerodynamic characteristics of the blades, resulting in a decrease in power generation efficiency and a potential increase in the load on the blade structure. The first task is to calculate and warn the ice thickness of the wind turbine blades. In this scenario, the task is to use the changes in the blade's own weight and inclination to infer the thickness of the ice layer attached to the blade surface, so as to promptly remind the operation and maintenance personnel to clean or adjust the blade posture.

[0051] The first task data includes:

[0052] Continuous weighing data: Install high-precision strain gauges or weighing devices on the wind turbine hub or blade root. When the blades are covered with ice and gain weight, the overall load changes will be reflected through the strain device.

[0053] Ice thickness labeling data: In the initial stage, some known ice conditions (which can be calibrated through drone inspections or ground observations) are recorded to establish a labeling data set for subsequent model training and verification.

[0054] Mission area of ​​the first object:

[0055] The critical ice attachment area of ​​a wind turbine blade is usually the leading edge and windward surface of the blade where ice is more likely to form.

[0056] The second task data includes:

[0057] Inclination data: Inclination sensors are installed at the root or middle of the blade to monitor blade attitude changes. When the center of gravity of the blade shifts due to uneven ice coverage, the inclination data may change significantly, providing auxiliary information for ice thickness estimation.

[0058] The specific collection steps include installing strain gauges on the hub or root of the wind turbine to ensure that they can sensitively sense changes in the overall load of the blade. Install inclination sensors at appropriate locations on the blades to monitor blade posture fine-tuning and changes. Collect weighing and inclination data in real time through the SCADA system or local data recorder. In the early stages, compare drone inspections or manual observation data to determine the marking information of the ice thickness of the blades at the corresponding moment and include it in the data set. Compare the inclination with the weighing data, delete inconsistent data points, and standardize and normalize the remaining data.

[0059] S2: Use the first task data and the second task data to perform mutual verification and perform a first pre-processing on the data.

[0060] In the present invention, the first pre-processing includes data cleaning and data standardization. The weighing data and the inclination data are used to verify each other, clean the data, and correct the errors in the data. The weighing data and the inclination data are compared and analyzed to verify the consistency and accuracy of the data. Identify and correct possible errors or outliers in the data. Use data cleaning technology to remove noise and interference in the data to ensure the accuracy and reliability of the data. Establish a standard operating procedure for data cleaning and correction to ensure the consistency and standardization of data processing.

[0061] It should be noted that the first pre-processing includes but is not limited to data cleaning and data standardization, and data completion and data smoothing processing may also be added.

[0062] In an optional embodiment of the present invention, the first pre-processing includes data completion and data smoothing processing in addition to data cleaning and data standardization.

[0063] Furthermore, data completion is applicable to scenarios where, during the data collection process, data may be missing in some time periods due to sensor failure, data transmission interruption, or environmental interference. In order to ensure the integrity of the data set, the missing data needs to be completed.

[0064] The processing steps include checking the time series of the collected weighing data and inclination data to identify the location and range of missing data. If the number of missing data points is small and the data changes are relatively stable, linear interpolation can be used to fill in the missing values. Linear interpolation calculates reasonable interpolation values ​​by using the data trends before and after the missing points. If the number of missing data points is large or the data changes have nonlinear characteristics, use multivariate interpolation methods (such as Lagrange interpolation or spline interpolation). If the missing data involves multiple parameters (such as both weighing and inclination are missing), train a machine learning model based on historical data (such as random forest or LSTM) to predict the missing values.

[0065] The model input is the historical value and time characteristics of the relevant variables, and the output is the prediction of the missing value. Compare the completed data with the trend of the adjacent points to check whether it conforms to the overall distribution and change trend of the data. If there is a deviation in the completed data, it is necessary to adjust the completion model or re-analyze the data characteristics.

[0066] Furthermore, data smoothing is applicable to scenarios where the collected weighing data and inclination data may fluctuate or change abnormally in a short period of time because the sensor may be disturbed by the external environment (such as vibration, wind, temperature change, etc.). Data smoothing can effectively eliminate noise and improve the stability and availability of data.

[0067] The processing steps include continuity analysis of the collected weighing and inclination data to identify sharp fluctuations or abnormal jump points in the data. Define a sliding window (the past 5 data points), calculate the average value of the data in the sliding window at each time point, and use it as the smoothing value of the current point. The size of the sliding window needs to be selected according to the frequency and characteristics of the data fluctuations. Smaller windows are suitable for capturing rapidly changing trends, while larger windows are suitable for smoothing long-term trends. Perform wavelet decomposition on the time series data to decompose it into components of different frequencies. Soft threshold processing is performed on the high-frequency component (noise) to reduce its impact while retaining the low-frequency component (key trend).

[0068] The processed data are restored to a smooth sequence through wavelet reconstruction.

[0069] Check the smoothed data against the original data to ensure that the smoothing process did not introduce new anomalies or change the actual trend of the data.

[0070] If the data fluctuates greatly in certain time periods, the size of the sliding window or the wavelet threshold is dynamically adjusted to meet the smoothing requirements of different scenarios. The smoothed data is saved as new input data for subsequent ice thickness calculation.

[0071] S3: Collect environmental data and build a data set by combining the pre-processed data.

[0072] Select appropriate meteorological sensors to obtain meteorological data such as temperature, humidity, etc. Integrate meteorological data with weighing data and inclination data to construct a multivariate time series data set.

[0073] Ensure the time synchronization between sensor data for effective data fusion and analysis. Establish a data fusion algorithm to normalize the information from different data sources and integrate them into a unified data set.

[0074] The specific steps of normalization are as follows:

[0075] Determine the minimum and maximum values ​​for each feature in the dataset. Normalize each data point for each feature by multiplying the normalized value by the desired width of the range, usually 1, and then adding the minimum value of the range.

[0076] In this way, data of any numerical range can be converted to the interval [0,1], which helps to balance the scale effects of different features in machine learning algorithms.

[0077] S4: Build a neural network model to complete the calculation of the task.

[0078] Design the deep learning network structure, taking into account the data characteristics and the training effect of the model. Input the multivariate time series data set into the deep learning network to train the model. Optimize the model parameters to improve the accuracy and generalization ability of the model. Perform cross-validation and model evaluation to ensure the effectiveness and reliability of the model.

[0079] Based on the trained neural network, the ice thickness of the distribution line is calculated according to the collected weighing, inclination and meteorological data. The process includes the following steps:

[0080] The weighing, inclination and meteorological data collected in real time are input into the trained neural network model.

[0081] The neural network model is used to process and calculate data and predict the ice thickness of distribution lines.

[0082] Evaluate and verify the calculation results to ensure their accuracy and reliability.

[0083] Establish a real-time monitoring system for the thickness of ice covering distribution lines to achieve continuous monitoring and early warning of ice coverage.

[0084] The computer device may be a server. The computer device includes a processor, a memory, an input / output interface (Input / Output, referred to as I / O) and a communication interface. The processor, the memory and the input / output interface are connected via a system bus, and the communication interface is connected to the system bus via the input / output interface. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the computer device is used to store data cluster data of the power monitoring system. The input / output interface of the computer device is used to exchange information between the processor and an external device. The communication interface of the computer device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, a method for calculating ice thickness based on continuous weighing and inclination data is implemented.

[0085] A person of ordinary skill in the art can understand that all or part of the processes in the above-mentioned embodiment method can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to the memory, database or other medium used in the embodiments provided in the present application can include at least one of non-volatile and volatile memory. Non-volatile memory may include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory may include random access memory (RAM) or external cache memory, etc. As an illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM). The database involved in each embodiment provided in this application may include at least one of a relational database and a non-relational database. Non-relational databases may include distributed databases based on blockchains, etc., but are not limited to this. The processor involved in each embodiment provided in this application may be a general-purpose processor, a central processing unit, a graphics processor, a digital signal processor, a programmable logic device, a data processing logic device based on quantum computing, etc., but are not limited to this.

[0086] Example 2 is an embodiment of the present invention, which provides a method and system for calculating ice thickness based on continuous weighing and inclination data. In order to verify the beneficial effects of the present invention, scientific demonstration is carried out through simulation experiments.

[0087] In order to verify the performance superiority of a method for calculating ice thickness based on continuous weighing and inclination data, a section of an actual transmission line was selected as the test object. In the test, the existing technology and the method invented by our company were used to monitor and calculate the ice thickness of the line, and the results were compared and analyzed.

[0088] The prior art is as follows:

[0089] A single type of sensor (such as a weighing sensor or an inclination sensor) is installed in key areas of the transmission line. The line weight or inclination data is collected separately, and the ice thickness is estimated using a linear regression model. The data processing stage only performs simple mean processing on the collected data, and does not clean up the noise data and outliers. No joint analysis is performed with meteorological data, and only relies on direct inference of single sensor data.

[0090] The implementation process of our invention is as follows:

[0091] Weighing sensors and inclination sensors are installed at key locations of the transmission line, and meteorological sensors (collecting temperature, humidity and other data) are deployed at the same time. The icing data output by the sensors is collected, including the weight change and inclination change of the line.

[0092] The collected weighing and inclination data are cross-validated to exclude abnormal data, and data cleaning and standardization are performed. Abnormal data is smoothed using the sliding window filtering method to ensure data stability and consistency.

[0093] The pre-processed weighing data, inclination data and meteorological data are integrated into a multidimensional data set, and the ice thickness is trained and calculated using a neural network model based on deep learning. The model input includes multiple time series data, and the model output is a dynamic prediction value of ice thickness.

[0094] Through real-time transmission, the predicted ice thickness is output, and the error range of the data is recorded and compared with the actual situation to optimize the model.

[0095] The test period was 10 consecutive days, and the daily changes in ice thickness were recorded to compare the accuracy, error range and response time of the two methods. The experimental results are shown in Table 1.

[0096] Table 1 Experimental results

[0097]

[0098]

[0099] It is obvious from the test results that our invention has significant advantages over the existing technology in terms of ice thickness calculation accuracy, error control and data processing capabilities:

[0100] The existing technology relies on a single data source and is limited by the accuracy of sensor collection. The calculation error range of ice thickness is large, reaching 0.52.0cm. However, our invention greatly reduces the error range by integrating weighing, inclination and meteorological data, stabilizing it between 0.20.7cm, and improving the accuracy by more than 60%.

[0101] The existing technology does not deal with abnormal data well and directly uses raw data with noise, which leads to large fluctuations in calculation results. Our invention significantly improves data quality through cross-validation and data cleaning, making the results more consistent and stable.

[0102] Existing technologies use linear regression or simple empirical formulas, which cannot adapt to the complex nonlinear relationship between data. Our invention uses a deep learning model to fully explore the correlation between multiple variables, making the calculation results under complex meteorological conditions closer to reality.

[0103] The data in the table show that under high humidity and low temperature conditions (such as -4℃ / 65% humidity), the error of the existing technology increases significantly (up to 2.0cm). This shows that its method cannot effectively cope with extreme climates. Our invention has better adaptability because it integrates meteorological data and adopts a real-time dynamic adjustment model strategy.

[0104] Embodiment 3 is an embodiment of the present invention, and further includes an ice thickness calculation system based on continuous weighing and inclination data, specifically:

[0105] A data collection module, collecting first task data of a first object and collecting second task data of a task area of ​​the first object;

[0106] A preprocessing module, using the first task data and the second task data to perform mutual verification and perform a first preprocessing on the data;

[0107] Constructing a data set module, collecting first environment data, and combining the pre-processed data to construct a data set;

[0108] The computing module builds a neural network model to complete the calculation of the task.

[0109] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

Claims

1. A method for calculating ice thickness based on continuous weighing and inclination data, characterized in that: include: Collecting first task data of a first object, and collecting second task data of a task area of ​​the first object; Using the first task data and the second task data to perform mutual verification, and performing a first pre-processing on the data; Collect environmental data and build a data set by combining pre-processed data; Build a neural network model to complete the calculation of the task.

2. The ice thickness calculation method based on continuous weighing and inclination data according to claim 1, characterized in that: The first task data includes installing a collection device in the first object to collect the first task data.

3. The ice thickness calculation method based on continuous weighing and inclination data according to claim 2, characterized in that: The mission area of ​​the first object is an ice-covered area of ​​the first object, and a collection device is installed to collect second mission data.

4. The ice thickness calculation method based on continuous weighing and inclination data according to claim 3, characterized in that: The mutual verification includes comparing and analyzing the first task data and the second task data, verifying the consistency of the data, and deleting inconsistent data; The first pre-processing is to pre-process the retained consistent data to improve data accuracy and standardization.

5. The ice thickness calculation method based on continuous weighing and inclination data according to claim 4, characterized in that: The constructing of the data set includes integrating the first environmental data and the pre-processed data to construct the data set.

6. The ice thickness calculation method based on continuous weighing and inclination data according to claim 5, characterized in that: The first object includes but is not limited to a power transmission line, and the first task data includes ice weight and ice thickness marking data on the first object; The second mission data includes the inclination data and ice thickness mark data of the mission area of ​​the first object; The first pre-processing includes but is not limited to data cleaning and standardization; Environmental data includes meteorological data.

7. The ice thickness calculation method based on continuous weighing and inclination data according to claim 6, characterized in that: The step of constructing a neural network model to complete the calculation of the first task includes constructing a deep learning network and training the model using the data set; Based on the trained neural network model, the newly collected first task data and second task data are input to complete the calculation of the task.

8. An ice thickness calculation system based on continuous weighing and inclination data using the method according to any one of claims 1 to 7, characterized in that: A data collection module, collecting first task data of a first object and collecting second task data of a task area of ​​the first object; A preprocessing module, using the first task data and the second task data to perform mutual verification and perform a first preprocessing on the data; Constructing a data set module, collecting first environment data, and combining the pre-processed data to construct a data set; The computing module builds a neural network model to complete the calculation of the task.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 7 are implemented.

Images